T cells play a critical role in the host response to Mtb infection, and are their antigen specificity forms the basis of widely used clinical immunodiagnostic tests. MHC-restricted effector memory T cells protect the host from progression to active TB disease, but the particular effector functions, which mediate control are unknown. In a well characterized human patient cohort in Lima, Peru, we will determine the association of active TB disease with 400 transcripts measured in highly purified effector memory T cells. Transcripts associated with lack of disease progression will be validated by PCR and protein measurements to define gene products that can be measured as surrogates for protection from TB progression and targets for immunotherapy development. Nearly all current technology development efforts related to adjuvant formulation, vaccine design and immunodiagnosis focus on MHC antigen presenting molecules. However, recent studies show that non-classical CDIb and MRI proteins present mycobacterial lipids and metabolites to T cells in TB disease. Emphasizing new ex vivo methods and CD1 tetramers, we will measure the relationship of expansion of lipid- and metabolite-specific GEM T cells and MAIT cells during human and guinea pig infection and relapse. In particular, we propose to (a) measure GEM T cell expansion ex vivo during acute human tuberculosis infection (b) comparative nanostring profiling of effector functions of MAIT cells and GEM T cells, and (c) detect CDIb-restricted T cells in guinea pigs using CDIb tetramers. These translational studies seek to establish the first tractable small animal model of in vivo CDIb function and detect a causal relationship of infection with invariant T cell activation. Bypassing the genetic complexities of human MHC proteins, activation of invariant T cells by lipids and metabolites offers potentially more uniform outcomes that could be detected or modulated with lipids and vitamin metabolites (Project 4).